I agreed that SNES_DIVERGED_OSCILLATORY would even better. SNES_DIVERGED_USER will be still useful in the future in case we have some user-designed divergence criteria.
Thanks, Fande On Thu, Dec 3, 2020 at 1:07 PM Matthew Knepley <knep...@gmail.com> wrote: > On Thu, Dec 3, 2020 at 2:27 PM Alexander Lindsay <alexlindsay...@gmail.com> > wrote: > >> Would you guys support addition of a `SNES_DIVERGED_USER` to >> `SNESConvergedReason`? We already have `SNES_LINESEARCH_FAILED_USER`, so I >> think there is some precedent. The motivator is here: >> https://github.com/idaholab/moose/pull/16377. None of the current >> options fit the reason we want to error, and this is truly a user criterion >> for considering the nonlinear solve diverged. >> > > We can definitely add something. Why would it not be > SNES_DIVERGED_OSCILLATORY? > > Thanks, > > Matt > > >> Alex >> > -- > What most experimenters take for granted before they begin their > experiments is infinitely more interesting than any results to which their > experiments lead. > -- Norbert Wiener > > https://www.cse.buffalo.edu/~knepley/ > <http://www.cse.buffalo.edu/~knepley/> >
